[phenixbb] fixing a partial solution of mol. rep.

Maia Cherney chern at ualberta.ca
Tue Jan 6 16:07:35 PST 2009

Hi Randy,
thank you for your reply. We are using phenix.automr. We managed to fix 
a "core" ensemble and get positions for the 6 N-terminal domains 
(ensemble 2). The relationships between the core and N-domains are not 
clear yet. Here are some problems. 1. The N-domains are not the closest 
ones to the core. Some of them are very far. 2. The origin of the 
solution is moved relative to the fixed ensemble. It makes it difficult 
to compare different solutions. Is there a way to determine which 
N-domains are the same from different solutions?


Randy Read wrote:
> Hi Maia,
> If you're running AutoMR from the command-line wizard, you can specify 
> fixed components with the various fixed_* variables.  You can find an 
> example for fixing beta-lactamase and searching for BLIP in the 
> documentation 
> (http://www.phenix-online.org/documentation/automr.htm).  It's 
> possible to specify that you have six fixed components (for the six 
> copies of the monomers making up the core hexamer), but it might be 
> easier to specify a "core" ensemble containing the partial solution 
> for the core hexamer, and then specify it as a fixed ensemble with 
> orientation angles and translation vector both of  0,0,0.
> You can also do this in the AutoMR wizard run as a GUI, by choosing 
> the various fixed_* variables in the "Choose variable to set" pulldown.
> The third option (though this may not be the right forum to mention 
> it!) is to run Phaser from the ccp4i interface.  AutoMR is much easier 
> for more straightforward problems, but the ccp4i interface is still 
> more flexible for getting at advanced features of Phaser in difficult 
> cases.  We're working to improve the Phenix interface for the next 
> release of Phenix.  Anyway, in the ccp4i interface, you just provide a 
> .sol file containing the 6 SOLU 6DIM... lines defining the core 
> hexamer as a fixed partial solution, then tell Phaser to look for 
> copies of the other monomer.
> In a difficult case like this, it's possible that one or two of the 
> copies of the second component will give a clear solution, which would 
> show the relationship between the monomers in the core and monomers of 
> the second molecule.  Then you could try solving the structure by 
> making an alpha+beta object (where alpha is a monomer in the core, and 
> beta is the second type of monomer) and looking for six copies of 
> that, or superimpose copies of the alpha+beta object on copies of 
> alpha still lacking a companion.
> Good luck!
> Randy
> On 24 Nov 2008, at 00:33, chern wrote:
>> I am doing a molecular replacement.
>> There are two types of protein molecules in the crystal. One type is 
>> forming a hexameric core; there are 6 molecules of the second type 
>> that I need to locate.
>> I found a solution for the hexameric core, but the other 6 molecules 
>> are difficult to find.
>> 1. Can I fix the solution for the core, so that I don't have to find 
>> it every time. The other inconvenience with that is that the origine 
>> is different for every new run. If I fix the hexamer, then I fix the 
>> origine. It's important to me to compare different runs of MR.
>> 2. Some of the six other molecules that MR finds are not the closest 
>> to the core. How can I request to output the closest molecules?
>> Maia
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb at phenix-online.org
>> http://www.phenix-online.org/mailman/listinfo/phenixbb
> ------
> Randy J. Read
> Department of Haematology, University of Cambridge
> Cambridge Institute for Medical Research      Tel: + 44 1223 336500
> Wellcome Trust/MRC Building                   Fax: + 44 1223 336827
> Hills Road                                    E-mail: rjr27 at cam.ac.uk
> Cambridge CB2 0XY, U.K.                       
> www-structmed.cimr.cam.ac.uk
> ------------------------------------------------------------------------
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://www.phenix-online.org/mailman/listinfo/phenixbb

More information about the phenixbb mailing list